Sage Journals: Discover world-class research

Abstract

Objectives:

The aim of this experimental study was to investigate the effects of mad honey (grayanotoxin, GTX), used in complementary medicine for a variety of purposes besides being food, on pain thresholds in normal mice as model for acute pain and diabetic mouse as model for neuropathic pain.

Methods:

Hind paw withdrawal pain threshold to thermal stimulus was measured with a plantar analgesia meter in a mice model using healthy intact animals for acute pain and streptozotocin-induced diabetic animals for chronic neuropathic pain. Time and dose-dependent effects of intraperitoneally (i.p.) administered GTX were investigated in both acute and neuropathic pain.

Results:

In the acute pain model, administration of GTX caused a dose- and time-dependent marked increase in the pain latency values. In diabetic mice, which had markedly increased threshold to pain, GTX (0.1 mg/kg, i.p.) restored the mean pain latencies by decreasing from the pre-GTX treatment values of 3.2 ± 0.6 to 3.0 ± 0.9s at 10 min, 3.2 ± 0.6s at 20 min, 3.4 ± 0.6s at 30 min, 2.6 ± 0.5s at 60 min and 2.4 ± 0.6s (p < 0.05) at 100 min.

Conclusion:

The results from this experimental study indicate that GTX exhibits significant analgesic activity and has potential benefits against painful diabetic neuropathy. This is compatible with the widespread use of GTX containing mad honey for alleviating pain. Further studies involving long-term applications are needed for a more decisive conclusion regarding the usefulness of GTX as an analgesic, especially in the treatment of painful neuropathy.

Keywords

Mad honey grayanotoxin pain diabetic neuropathy mice streptozotocin

Introduction

Consumption of grayanotoxin (GTX)-containing honey products causes rarely fatal emergency of “mad honey” poisoning. Although cases have been reported around the world, mad honey poisoning is common in the eastern Black Sea region of Turkey.^1,2 When bees feed on the specific types of flower of rhododendron, this honey is produced and its ingestion resulting in poisoning has been called mad honey poisoning. Despite being toxic, mad honey is traditionally produced and marketed as an alternative medicine for a variety of health benefits including blood glucose regulation,³ gastrointestinal pain and disturbances, hypertension, pain relieving and is believed to be a sexual stimulant.⁴

GTXs are diterpenes, polyhydroxylated cyclic hydrocarbons that do not contain nitrogen.⁵ They occur in the nectar, pollen and other plant parts in some members of the family Ericaceae such as Rhododendron L. The genus Rhododendron is represented by six species in Flora of Turkey.^5,6 In Turkey, the commonly found toxic Rhododendron species are Rhododendron luteum L. and Rhododendron ponticum L.^6,7 There are at least 60 different GTXs,⁸ but the primary toxic compounds are GTXs I and III.^8,9

GTX has been shown to modify a region of the sodium (Na⁺) channel by binding to it, which is involved in the voltage-dependent activation and inactivation, and Na⁺ channel gating.¹⁰ Voltage-dependent Na⁺ channels play a major role in neuronal excitability characteristics. The effect of changes in various ion channel subtypes in different pain models has been revealed, and there is evidence that various types of ion channels are involved in nociception and heightened pain sensation. In essence, it has been hypothesized for several years that voltage-dependent Na⁺ channels play specialized roles in nociception and pain mechanisms.¹¹

In Black Sea region, mad honey has long been widely used, through both oral ingestion and topical application, for its health benefits in a variety of painful conditions including pain from burns, gastrointestinal pain, rheumatic or dental pain, common cold and edema.¹² But as yet there is no scientific evidence of the effectiveness or even safety of this old folk medicine as an analgesic. Hence, in this experimental study, we tested the effects of GTX on pain thresholds in normal mice (model for acute pain) and in a painful diabetic neuropathy model of pain.

Methods

Experimental design

All protocols were reviewed and approved by the Institutional Ethics Committee for Animal Experimentation.

In total, 85 adult male BALB-C mice, weighing an average of 25–30 g, were obtained from Karadeniz Technical University Medical School Experimental Research Center (Trabzon, Turkey). The animals were housed under standard light/dark condition (12-h light and 12-h dark) at constant temperature (21 ± 1°C) and humidity (55 ± 5%) with unrestricted access to standard laboratory animal chow and water. Study groups were randomized, and of the total animals, 30 mice served as normoglycemic group for acute pain investigations, while experimental diabetes was induced in 45 mice, and 10 mice served as an intact control group.

Study protocol

Hind paw withdrawal nocifensive response latency was determined by focusing a beam of thermal radiant heat to the plantar surface of the hind paw through glass floor using a plantar analgesia meter (MAY PWAM 0903 Plantar Test, Ankara, Turkey) with appropriate restrainer dimensions for two mice.^13,14 The nocifensive response on this system is considered to result from a coordinated activity of central and peripheral mechanisms.¹⁵ In brief, mice were placed in Plexiglass boxes positioned on a glass surface. Animals were allowed to acclimate to the hot plate for a period of 1 week prior to the experiment. The movable radiant heat source was targeted at the plantar area of the hind paw and activated with a light beam intensity, chosen in preliminary experiments to give baseline latencies from 5 to 6s in control mice. The moment mice withdrew their paws, the system was terminated automatically so “thermal pain threshold” latency was automatically determined in seconds (using a digital indicator), and measurements were then transferred to computer.

Sign of paw withdrawal, shaking, licking of the paw or jump response were taken as pain response. A cutoff duration of 6s was used to prevent soft tissue damage in the absence of response. Thermal nociceptive threshold values were measured thrice for each mouse, at intervals of at least 1 min, and the average values were calculated. Measurements were performed and evaluated by the same analyst, blind to the study protocols.

Diabetes was induced by intraperitoneal (i.p.) injection of streptozotocin (STZ, Sigma, Deisenhofen, Germany) at a dose of 200 mg/kg, according to the previously described protocols.^13,14 STZ is dissolved in a 0.4-ml (0.1 M) Na⁺ citrate buffer (pH 4.5) and is given as a single dose by i.p. injection. Mice blood glucose levels were investigated and recorded using glucometry from the tail vein. After 1 week, blood was again taken from the tail vein, and mice with a replete blood glucose level of >300 mg/dl, measured using a glucometer (Bioland Glucometer, Kowloon, Hong Kong), were regarded as diabetic.^16,17 Experiments were conducted 4 weeks after STZ injection, and those animals with a serum glucose level above 400 mg/dl were used for data analysis.

The GTX (Sigma, Deisenhofen, Germany) doses tested were 0.05 mg/kg, 0.01 mg/kg and 0.02 mg/kg for both normoglycemic mice and diabetic mice. GTX injections were given intraperitoneally in a single dose; control animals received an equivalent volume of physiological saline. Animals in each group were individually exposed to plantar analgesia test and mice were tested for heat withdrawal latency at baseline (0 min) and at 10, 20, 30, 60 and 100 min after GTX injection. Normoglycemic animals for acute pain consisted of 10 animals, and mice with diabetic neuropathy used for neuropathic pain consisted of 15 animals in each dose group.

Data analysis

Data are expressed as mean ± SEM. Pain threshold values were determined and analyzed by Kruskal–Wallis one-way analysis of variance followed by a pairwise comparison between control, before and after GTX for each GTX-treated group using Dunnett’s t test on the ranked data. A value of p < 0.05 was considered statistically significant. All statistical analyses were performed by using SPSS 16.0 (Chicago, Illinois, USA).

Results

Compared with the respective baseline withdrawal threshold, a single dose of GTX-injection (0.05, 0.1 and 0.2 mg/kg GTX) caused increase in pain threshold latency in a time and dose-dependent manner in normal mice. However, injection of physiological saline and vehicle of GTX had no significant effect on the withdrawal threshold to noxious heat stimuli.

Administration of 0.05 mg/kg GTX caused the heat withdrawal threshold to increase from 2.0 ± 0.3s to 2.5 ± 0.7s, 2.8 ± 0.6s, 3.0 ± 0.5s, 3.0 ± 0.6s and 3.1 ± 0.6s when measured at 10, 20, 30, 60 and 100 min after injection, respectively (Figure 1). A dose of 0.1 mg/kg GTX increased the heat withdrawal latency from 2.1 ± 0.2s to 3.5 ± 0.7s (p < 0.01), 3.9 ± 0.5s (p < 0.01), 3.2 ± 0.7s (p < 0.05), 3.3 ± 0.6s (p < 0.05) and 3.0 ± 0.5s (p < 0.05) at 10, 20, 30, 60 and 100 min after injection, respectively (Figure 2). Administration of the highest dose of GTX tested (0.2 mg/kg) increased the threshold latency in normoglycemic mice. Significant effect was observed at the first measure carried out 10 min after GTX injection, and this time dependently escalating effect remained consistent over the entire test duration (Figures 1 and 2).

Figure 1.

Time-dependent effects of 0.05 mg/kg grayanotoxin on pain threshold in normoglycemic mice as model for acute pain subject to plantar test.

Figure 2.

Time-dependent effects of 0.1 mg/kg grayanotoxin on pain threshold in normoglycemic mice as model for acute pain subject to plantar test.

The effects of GTX were also tested in mice model of diabetic neuropathy. Induction of diabetes led to a significantly increased pain threshold values as compared to their baseline and control group values, revealing the diabetic neuropathy development. This response was dose and time dependently reversed by GTX administration. But no significant latency modification was observed after injection of physiological saline along the 100-min testing period for the diabetic mice. Administration of 0.05 mg/kg GTX caused the mean heat withdrawal threshold values to decrease from 3.0 ± 0.5s to 2.9 ± 0.6s, 3.1 ± 0.7s, 3.0 ± 0.6s, 2.9 ± 0.6s and 2.4 ± 0.5s (p < 0.05) when measured at 10, 20, 30, 60 and 100 min after injection, respectively (Figure 3). A dose of 0.1 mg/kg GTX increased the heat withdrawal latency from 3.2 ± 0.6s to 3.0 ± 0.9s, 3.2 ± 0.6s, 3.4 ± 0.6s, 2.6 ± 0.5s and 2.4 ± 0.6s (p < 0.05) at 10, 20, 30, 60 and 100 min after injection, respectively.

Figure 3.

Time-dependent effects of 0.05 mg/kg grayanotoxin on pain threshold in streptozotocin-induced hyperglycemic mice as model for diabetic neuropathy subject to plantar test.

Administration of the highest dose of GTX tested (0.2 mg/kg) increased the threshold latency in diabetic mice from 3.2 ± 0.4s to 2.9 ± 0.7s, 3.0 ± 0.8s, 3.2 ± 0.7s, 2.6 ± 0.8s and 2.4 ± 0.5s (p < 0.05) at 10, 20, 30, 60 and 100 min after injection, respectively (Figure 4).

Figure 4.

Time-dependent effects of 0.2 mg/kg grayanotoxin on pain threshold in streptozotocin-induced hyperglycemic mice as model for diabetic neuropathy subject to plantar test. Data are expressed as mean ± SEM compares to just prior to the GTX injection values (0 min) and shown in figure are basal treatment (measured just prior to the levetiracetam injection at 0 min) and GTX treatment (measured at indicated time points after GTX injections latencies, seconds) for the nociceptive reaction.

In painful diabetic neuropathy model, post injection effect of GTX was observed at the first measure carried out at 10 min after GTX injection, and this time-dependent effect being significant only at the 100th min.

Discussion

Overcoming and ameliorating pain are the major subjects that have preoccupied physicians throughout medical history. Medicine has made great progress in the treatment of acute pain, and various alternative treatment modalities have been produced. But neuropathic pain still remains a clinical challenge. Neuropathic pain affects approximately 1% of the population and accompanies many types of neuropathy. The most frequent cause is diabetes.¹⁸

The majority of pharmacological treatments have focused on reducing fast Na⁺ channels and ectopic discharges that are held responsible for neuropathic pain. These defects can be rectified either with Na⁺ channel antagonists or drugs that inhibit spinal neurotransmission.¹⁹ The existing evidence does not support the superiority of one drug over another.

Biological toxins affecting voltage-dependent Na⁺ channels are used to illuminate the molecular mechanism of the Na⁺ channel opening process. GTX first binds to the open part of the Na⁺ channel, then the modified Na⁺ channel is exposed to an inactivation process and the activation potential in the modified Na⁺ channel is shifted toward hyperpolarization.¹

It is well known that gamma-amino-n-butyric acid (GABA) has alleviating effects on neuropathic pain.²⁰ Sitges has reported that GTX stimulates GABA release from isolated nerve terminals.²¹ Although we have no direct evidence, the release of GABA and/or the modulation of Na⁺ channels are among the plausible mechanisms of action of GTX observed in this study.

The use of rodents in the model of painful neuropathy employed in this study permits the evaluation of both electrical and neurochemical activities of nervous system and also behavioral responses to sensory stimuli. There is an early deceleration of nerve transmission in diabetic rats and mice. The physiological, neurochemical and behavioral changes in rodent painful neuropathy are mainly due to hyperglycemia. Behavioral studies in diabetic rats are conducted with hyperalgesia and allodynia tests. Exposure of the tail or paw to heat can be used to determine animals’ extremity withdrawal durations, hyperalgesia and hypoalgesia.^22,23

In our study, GTX in different doses in normoglycemic rat has different levels of efficacy and raises pain threshold in a statistically significant manner. This suggests that GTX can have a preventive effect on acute pain. The fact that the acute effect began from 10 min after GTX use and persisted until the 100th min is instructive in terms of the efficacy period. According to our findings, latency increase is greater in the early period, and the increase then gradually declines. The effect of GTX on pain may be thought of as exhibiting time-dependent differences. Neuropathic pain latency levels decreased to an average 2.4s approached very close to the prediabetic levels.

Limitations

The fact is that neuropathic pain groups were not suitable for our chronic pain model, but on the other hand, the latencies of neuropathic mice came close to the preneuropathic values. This suggested GTX has an ameliorating effect on neuropathy. GTX’s reduction in latency in neuropathic pain suggested it may be useful in the treatment of diabetic neuropathy. Further studies involving long-term GTX administration in mice with induced neuropathy are needed to confirm these data.

Conclusions

In conclusion, results from this experimental study indicate that GTX exhibits significant analgesic activity, supporting the widespread use of GTX-containing mad honey in complementary medicine for pain relief. Further studies involving long-term clinical applications are needed for a more decisive conclusion regarding the usefulness of GTX as an analgesic, especially its potential benefits against painful diabetic neuropathy.

Table 1.

Dose- and time-dependent effects of GTX on radiant heat-induced nociceptive threshold latency values (mean ± SEM) measured at different time points before and after intraperitoneal GTX administration in naive mice of acute pain model.

Groups	Pain latency (sec) at different time points after GTX administration (min)
Groups	0	10	20	30	60	100
Control	2.4 ± 0.5	2.3 ± 0.3	2.5 ± 0.4	2.6 ± 0.5	2.5 ± 0.4	2.6 ± 0.5
Control	2.4 ± 0.5	2.3 ± 0.3	2.5 ± 0.4	2.6 ± 0.5	2.5 ± 0.4	2.6 ± 0.5
0.05 mg/kg GTX	2.0 ± 0.3	2.5 ± 0.7	2.8 ± 0.6	3.1 ± 0.6	3.0 ± 0.5	3.0 ± 0.6
0.05 mg/kg GTX	2.0 ± 0.3	2.5 ± 0.7	2.8 ± 0.6	3.1 ± 0.6	3.0 ± 0.5	3.0 ± 0.6
0.1 mg/kg GTX	2.1 ± 0.2	3.5 ± 0.7	3.9 ± 0.5	3.2 ± 0.7	3.3 ± 0.6	3.0 ± 0.5
0.1 mg/kg GTX	2.1 ± 0.2	3.5 ± 0.7	3.9 ± 0.5	3.2 ± 0.7	3.3 ± 0.6	3.0 ± 0.5
0.2 mg/kg GTX	2.4 ± 0.4	3.1 ± 0.6	3.0 ± 0.3	2.9. ± 0.5	3.0 ± 0.5	3.2 ± 0.7

GTX: grayanotoxin.

Table 2.

Dose- and time-dependent effects of GTX on nociceptive threshold latency values (mean ± SEM) measured at different time points before and after intraperitoneal GTX administration in diabetes-induced neuropathic pain model.

Groups	Pain latency (sec) at different time points after GTX administration (min)
Groups	0	10	20	30	60	100
Control	2.4 ± 0.5	2.3 ± 0.3	2.5 ± 0.4	2.6 ± 0.5	2.5 ± 0.4	2.6 ± 0.5
Control	2.4 ± 0.5	2.3 ± 0.3	2.5 ± 0.4	2.6 ± 0.5	2.5 ± 0.4	2.6 ± 0.5
0.05 mg/kg GTX	3.0 ± 0.5	2.9 ± 0.6	3.1 ± 0.7	3.0. ± 0.6	2.9 ± 0.6	2.4 ± 0.5
0.05 mg/kg GTX	3.0 ± 0.5	2.9 ± 0.6	3.1 ± 0.7	3.0. ± 0.6	2.9 ± 0.6	2.4 ± 0.5
0.1 mg/kg GTX	3.2 ± 0.6	3.0 ± 0.9	3.2 ± 0.6	3.4 ± 0.6	2.6 ± 0.5	2.4 ± 0.6
0.1 mg/kg GTX	3.2 ± 0.6	3.0 ± 0.9	3.2 ± 0.6	3.4 ± 0.6	2.6 ± 0.5	2.4 ± 0.6
0.2 mg/kg GTX	3.2 ± 0.4	2.9 ± 0.7	3.0 ± 0.8	3.2 ± 0.7	2.6 ± 0.8	2.4 ± 0.5
0.2 mg/kg GTX	3.2 ± 0.4	2.9 ± 0.7	3.0 ± 0.8	3.2 ± 0.7	2.6 ± 0.8	2.4 ± 0.5

GTX: grayanotoxin.

Footnotes

Conflict of interest

The authors declared no conflicts of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Gunduz

Turedi

Russell

Ayaz

. Clinical review of grayanotoxin/mad honey poisoning past and present. Clin Toxicol 2008; 46: 437–442.

Gunduz

Tatli

Turedi

. Geçmişten günümüze deli bal zehirlenmesi [mad honey poisoning from past to present]. Turk J Emerg Med 2008; 8: 46–49.

Oztasan

Altinkaynak

Akcay

. Effect of mad honey on blood glucose and lipid levels in rats with streptozocin-induced diabetes. Turk J Vet Anim Sci 2005; 29: 1093–1096.

Koca

. Poisoning by mad honey: a brief review. Food Chem Toxicol 2007; 45: 1315–1318.

Lampe

. Rhododendrons, mountain laurel, and mad honey. JAMA 1988; 259: 2009.

Yavuz

Ozel

Akkus

Erkul

. Honey poisoning in Turkey. Lancet 1991; 337: 789–790.

Gunduz

Bostan

Turedi

Nuhoglu

Patan

. Wild flowers and mad honey. Wilderness Environ Med 2007; 18: 69–71.

Terai

Osakabe

Katai

Sakaguchi

Narama

Matsuura

Preparation of 9-hydroxy grayanotoxin derivatives and their acute toxicity in mice. Chem Pharm Bull 2003; 51: 351–353.

Ascioglu

Ozesmi

Dogan

Ozturk

. Effects of acute grayanotoxin I administration on hepatic and renal functions in rats. Turk J Med Sci 2000; 30: 23–27.

10.

Maejima

Kinoshita

Seyama

. Distinct site regulating grayanotoxin binding and unbinding to D4S6 of Nav1.4 sodium channel as revealed by improved estimation of toxin sensitivity. J Biol Chem 2003; 278: 9464–9471.

11.

Cummins

Sheets

Waxman

. The roles of sodium channels in nociception: implications for mechanisms of pain. Pain 2007; 131: 243–257.

12.

Baytop

. Türkiye’de Bitkiler ile Tedavi, Gecmiste ve Bugün [Therapy with Medicinal Plants in Turkey, Past and Present]. 2nd ed. Istanbul, Turkey: Nobel Tip Basimevi, 1999, p. 275.

13.

Hargreaves

Dubner

Brown

Flores

Joris

. A new and sensitive method for measuring thermal nociception in cutaneous hyperalgesia. Pain 1988; 32: 77–88.

14.

Ozcan

Ayar

Canpolat

Kutlu

. Antinociceptive efficacy of levetiracetam in a mice model for painful diabetic neuropathy. Acta Anaesthesiol Scand 2008; 52: 926–930.

15.

Bannon

Malmberg

. Models of nociception: hot-plate, tail-flick, and formalin tests in rodents. Curr Protoc Neurosci 2007; 41: 8.9.1–8.9.16.

16.

Motyl

McCabe

. Streptozotocin, type I diabetes severity and bone. Biol Proced Online 2009; 11: 296–315.

17.

Huan

. Streptozotocin-induced diabetic models in mice and rats. Curr Protoc Pharmacol 2008; 40: 5.47.1–5.47.14.

18.

Mason

Wahab

. Extracellular matrix metabolism in diabetic nephropathy. J Am Sac Nephrol 2003; 14: 1358–1373.

19.

Boulton

. Treatment of symptomatic diabetic neuropathy. Diabetes Metab Res Rev 2003; 19: 16–21.

20.

Catterall

. Neurotoxins that act on voltage-sensitive sodium channels in excitable membranes. Ann Rev Pharmacol Toxicol 1980; 20: 15–42.

21.

Sitges

. Characterization of the effect of monensin on gamma-amino-n-butyric acid release from isolated nerve terminals. J Neurochem 1989; 53: 442–447.

22.

Fox

Eastwood

Gentry

Manning

Urban

. Critical evaluation of the streptozotocin model in the rat. Pain 1999; 81: 307–316.

23.

Shafik

. Effects of topiramate on diabetes mellitus induced by streptozotocin in rats. Eur J Pharmacol 2012; 684: 161–167.

Analgesic effects of mad honey (grayanotoxin) in mice models of acute pain and painful diabetic neuropathy

Abstract

Objectives:

Methods:

Results:

Conclusion:

Keywords

Introduction

Methods

Experimental design

Study protocol

Data analysis

Results

Discussion

Limitations

Conclusions

Footnotes

Conflict of interest

Funding

References